American Society of Civil Engineers


Selecting a Domestic Water Demand Prediction Model for Climate Change Studies


by M. Karamouz, (Research Professor, Polytechnic Institute of NYU, Brooklyn, NYU, Professor, School of Civil Engineering, University of Tehran, Tehran, Iran. E-mail: mkaramou@poly.edu), Z. Zahmatkesh, Ph.D., (Student, School of Civil Engineering, University of Tehran, Tehran, Iran. E-mail: zahrazahmatkesh@ut.ac.ir), and S. Nazif, Ph.D., (School of Civil Engineering, University of Tehran, Tehran, Iran. E-mail: saranazif@yahoo.com)
Section: Climate Change Symposium, pp. 1338-1346, (doi:  http://dx.doi.org/10.1061/41173(414)139)

     Access full text
     Purchase Subscription
     Permissions for Reuse  

Document type: Conference Proceeding Paper
Part of: World Environmental and Water Resources Congress 2011: Bearing Knowledge for Sustainability
Abstract: Water demand analysis is needed for design, operation and management of urban water supply systems. Rapid urbanization, economical and industrial developments, and growth of population especially in developing countries have resulted in increasing water deficiency. The problem is more intensified in urban areas with a high density of population and limited water supply resources. In addition some climatic and socio-economic changes will intensify the water limitations. All of these changes should be incorporated in future water supply and demand management. Due to data availability and the importance of demand modeling in water supply planning, different methods are developed for demand simulation. In this study, the relative performance of different artificial neural network (ANN) techniques such as Feed Forward Neural Networks (FFNN), Generalized Regression Neural Networks (GRNN) and regression based approaches for water demand prediction are investigated. Then a better model is selected for climate change studies in the central part of Iran. To incorporate climate change impacts on water demand, downscaled climatic data are used in the selected model to project future demand. The results of this study show the increasing gap between water supply availability and water demand. Demand side management in a more rigorous and integrated fashion should be employed to fill this gap.


ASCE Subject Headings:
Water demand
Climate change
Neural networks
Residential location